Rapid Service Deployment: An Operations Perspective

Abstract

The goal of Intelligent Networks (IN) is service To enable this, much emphasis has been placed on the development of an architecture that primarily focuses on (1) a Call Model to which all Service Logic Programs (SLPs) adhere, (2) the Service Logic Execution Environment (SLEE) on which the SLPs execute, and (3) the Service Creation Environment (SCE) on which the SLPs are created. However, the mechanisms and operations aspects of deploying such services have frequently been overlooked. Service deployment is defined as the complete process of loading the SLPs and relevant data onto the SLEE, and the creation and development of Operations Systems applications to administer the SLPs so that services can be made available to the customer, billed for, and maintained. In this talk, we first examine the critical role of operations support of a service, and based on our experiences in developing operations systems for GTE's IN testbed to achieve provisioning and deployment of services, we propose necessary requirements to truly achieve rapid service deployment. We introduce the concept of Service Package as the end product of the SCE and as the input to the deployment Operations Systems. We propose that the Service Package be standardized and so treated by the Operations Systems. Much efficiency can be attained by this as the deployment process is repeated for each new service. We also propose that all IN services adhere to a standardized Information Model (Management Information Base), akin to adherence to a standardized Call Model. In addition to being the underlying basis for constructing the Service Packages, the Information Model must be consistently followed in developing the SLPs, in preparing Operations Systems applications, and in deploying services.